Current issues in clinical equivalence trials.

A clinical trial designed to show that an experimental treatment E is similar to a control treatment S in a specified direction is a one-sided equivalence or similarity trial--in the terminology of the International Conference on Harmonisation, a non-inferiority trial (ICH, 1998). We design such a study to show that E is not worse than S (often an accepted or standard treatment) by as much as a pre-specified margin theta0. The quantity theta0 can be either a difference or ratio of an appropriate outcome in individuals treated with E and S. A critical issue is whether one can conclude from a non-inferiority trial that E is effective. Closely related is an appropriate choice of theta0, which should be substantially less than the estimated effect of S if available from previous studies; theta0 should also be acceptable to clinicians, either because of advantages of E or because a difference or ratio less than theta0 is considered unimportant clinically. Another possible approach for showing that E is effective is to estimate its effect compared with placebo from historical data. If previous studies that consistently show an effect of S are not available, alternative study designs should be considered. Findings of superiority or non-inferiority of E, when the study was planned to show the other, are possible and may be supportable. A finding that E is at the same time statistically significantly worse than S and "non-inferior" to S should not be a problem, if the criterion theta0 is appropriate and this possibility was considered in the protocol. Various sorts of non-adherence may make treatments appear similar, even if they are not. In particular, random non-adherence of study participants to the assigned treatment regimen may cause an intention-to-treat analysis to give a misleading result of similarity. Thus, maintaining a high degree of adherence to protocol is especially important in an equivalence or non-inferiority trial. Interim analysis does not present statistical problems in these trials; early stopping may not be wise in many cases, however, because strong interim evidence for non-inferiority may actually be an indicator that E is superior to S.

Allocation of subjects to test null relative risks smaller than one.

Allocating a proportion k'=1/(1+ radicalr(0)) of subjects to an intervention is a practical approach to approximately maximize power for testing whether an intervention reduces relative risk of disease below a null ratio r(0)<1. Furthermore, allocating k'(s), a convenient fraction close to k', to intervention performs nearly as well; for example, allocating k'(s)=3/5 for 0.5> or =r(0)>0.33,2/3 for 0.33> or =r(0)>0.17 and 3/4 for 0.17> or =r(0)> or =0.10. Both k' and k'(s) are easily calculated and invariant to alterations in disease rate estimates under null and alternative hypotheses, when r(0) remains constant. In examples that we studied, allocating k' (or k'(s)) subjects to intervention achieved close to the minimum possible sample size, given test size and power (equivalently, maximum power, given test size and sample size), for likelihood score tests. Compared to equal allocation, k' and k'(s) reduced sample sizes by amounts ranging from approximately 5.5 per cent for r(0)=0.50 to approximately 24 per cent for r(0)=0.10. These sample size savings may be particularly important for large studies of prophylactic interventions such as vaccines. While k' was derived from variance minimization for an arcsine transformation, we do not recommend the arcsine test, since its true size exceeded the nominal value. In contrast, the true size for the uncorrected score test was less than the nominal size. A skewness correction made the size of the score test very close to the nominal level and slightly increased power. We recommend using the score test, or the skewness-corrected score test, for planing studies designed to show a ratio of proportions is less than a prespecified null ratio r(0)<1.

Safety and immunogenicity of six acellular pertussis vaccines and one whole-cell pertussis vaccine given as a fifth dose in four- to six-year-old children.

OBJECTIVE: To evaluate the safety and immunogenicity of 6 different acellular pertussis vaccines combined with diphtheria and tetanus toxoids (DTaP) and with 1 licensed whole-cell pertussis vaccine (DTwP) as a fifth dose in children who had previously received the same DTaP, a different DTaP, or DTwP as primary and fourth-dose vaccinations. METHODS: Healthy 4- to 6-year-old children were enrolled at 5 National Institute of Allergy and Infectious Diseases Vaccine Treatment and Evaluation Units to receive a fifth dose of a DTaP or DTwP vaccine. All had been randomly assigned to receive 3 primary doses of DTaP or DTwP at 2, 4, and 6 months and a fourth-dose booster at 15 to 20 months of age as part of earlier National Institutes of Health multicenter acellular pertussis vaccine trials. Parents recorded the occurrence and magnitude of fever, irritability, and injection site redness, swelling, and pain for 3 days after vaccination. Sera obtained before and 1 month after the booster vaccination were analyzed by enzyme-linked immunosorbent assay for antibody to pertussis toxin, filamentous hemagglutinin, fimbriae, pertactin, and diphtheria and tetanus toxoid. Safety and/or immunogenicity data are reported for 317 children who received DTaP and 10 children who received DTwP. RESULTS: Fever and moderate or severe irritability were uncommon following the fifth dose of DTaP vaccine and were generally less frequent than following the fourth dose. However, for the DTaP vaccine groups, redness, swelling, and pain increased in prevalence compared with the fourth dose. The time course and frequency of reactions following DTaP vaccination were generally similar in children who received the same DTaP, a different DTaP, or DTwP for previous doses in the 5- dose series. No significant differences among the DTaP vaccines were detected in the occurrence of reactions, but the statistical power to detect differences was limited by sample size. Significant increases in antibodies directed against the included antigens were observed for all DTaP vaccines in paired pre- and post-fifth dose sera. Post-fifth dose antibody concentrations differed significantly among the DTaP vaccines. Some children in the study showed an antibody response to an antigen not reported to be in the DTaP vaccine. CONCLUSION: All the studied DTaP vaccines performed similarly with regard to reactions, whether given as a fifth sequential dose of the same vaccine, a mix of different DTaP vaccines in the 5-dose sequence, or after 3 DTwP and 1 DTaP vaccinations. Large injection site reactions occurred more frequently after the fifth dose of DTaP than after the previous 4 doses. A fifth dose of all DTaP vaccines induced an antibody response to those antigens contained in the vaccine. No DTaP was consistently most or least reactogenic or immunogenic.

Human papillomavirus, anal squamous intraepithelial lesions, and human immunodeficiency virus in a cohort of gay men.

Cross-sectional associations between human papillomavirus (HPV), anal squamous intraepithelial lesions (SIL), and human immunodeficiency virus (HIV) were studied in a cohort of gay men. HPV DNA was detected by generic and type-specific polymerase chain reaction (PCR) probes and hybrid capture assay (HC). HPV virus load was estimated by HC relative light unit (RLU) ratio. HPV prevalence, number of HPV types detected, and HC RLU ratios were each greater in HIV-positive than HIV-negative participants. Further, among HIV-positive men, HC RLU ratio was inversely associated with CD4 cell count. SIL was more frequent in HIV-positive participants, particularly those with a CD4 cell count <200/microL and was positively associated with HPV. Men with a high HC RLU ratio were nearly 3 times more likely to have SIL than were those both PCR- and HC-negative. These data support that HIV augments HPV-associated anal disease in this population.

Estimation of pertussis vaccine efficacy in the presence of covariates in three randomized trials.

We estimated efficacy of pertussis vaccines in three randomized controlled trials with adjustment for several baseline covariates: presence of one or more other children in the household, sex of the study child, and geographical area. Adjusted and unadjusted efficacy estimates differed only trivially. We also assessed the association of efficacy with time since vaccination and background pertussis incidence. The acellular vaccines, except for the two-component vaccine in the Stockholm trial, appeared to maintain their efficacy during two years of follow-up. In contrast, efficacy of a whole-cell vaccine decreased significantly in both the Stockholm and Italian trials. The relationship between efficacy and background incidence was not consistent across studies and vaccines.

Colonization with group B streptococci in pregnancy and adverse outcome. VIP Study Group.

OBJECTIVE: Our purpose was to study the association of cervicovaginal colonization with group B streptococci with pregnancy and neonatal outcome. STUDY DESIGN: A prospective study was conducted at seven medical centers between 1984 and 1989. Genital tract cultures were obtained at 23 to 26 weeks' gestation and at delivery. Prematurity and neonatal sepsis rates were compared between group B streptococci positive and negative women. RESULTS: Group B streptococci was recovered from 2877 (21%) of 13,646 women at enrollment. Heavy colonization was associated with a significant risk of delivering a preterm infant who had a low birth weight (odds ratio = 1.5, 95% confidence interval 1.1 to 1.9). Heavily colonized women given antibiotics effective against group B streptococci had little increased risk of a preterm, low-birth-weight birth. Women with light colonization were at the same risk of adverse outcome as the uncolonized women. Neonatal group B streptococci sepsis occurred in 2.6 of 1000 live births in women with and 1.6 of 1000 live births in women without group B streptococci at 23 to 26 weeks' gestation (p = 0.11). However, sepsis occurred in 16 of 1000 live births to women with and 0.4 of 1000 live births to women without group B streptococci at delivery (p < 0.001). CONCLUSIONS: Heavy group B streptococci colonization of 23 to 26 weeks' gestation was associated with an increased risk of delivering a preterm, low-birth-weight infant. Cervicovaginal colonization with group B streptococci at 23 to 26 weeks' gestation was not a reliable predictor of neonatal group B streptococci sepsis. Colonization at delivery was associated with sepsis.

A controlled trial of two acellular vaccines and one whole-cell vaccine against pertussis. Progetto Pertosse Working Group.

BACKGROUND: Concern about both safety and efficacy has made the use of whole-cell pertussis vaccines controversial. In some European countries, including Italy, the rate of vaccination against pertussis is low. METHODS: We conducted a double-blind trial in Italy in which infants were randomly assigned to vaccination at two, four, and six months of age with an acellular pertussis vaccine together with diphtheria and tetanus toxoids (DTP); a DTP vaccine containing whole-cell pertussis (manufactured by Connaught Laboratories); or diphtheria and tetanus toxoids without pertussis (DT). The acellular DTP vaccine was either one containing filamentous hemagglutinin, pertactin, and pertussis toxin inactivated with formalin and glutaraldehyde (SmithKline Beecham) or one with filamentous hemagglutinin, pertactin, and genetically detoxified pertussis toxin (Chiron Biocine). Pertussis was defined as 21 days or more of paroxysmal cough, with infection confirmed by culture or serologic testing. RESULTS: The efficacy of each vaccine, given in three doses, against pertussis was determined for 14,751 children over an average of 17 months, with cases included in the analysis if cough began 30 days or more after the completion of immunization. For both of the acellular DTP vaccines, the efficacy was 84 percent (95 percent confidence intervals, 76 to 89 percent for Biocine DTP and 76 to 90 percent for SmithKline DTP), whereas the efficacy of the whole-cell DTP vaccine was only 36 percent (95 percent confidence interval, 14 to 52 percent). The antibody responses were greater to the acellular vaccines than to the whole-cell vaccine. Local and systemic adverse events were significantly more frequent after the administration of the whole-cell vaccine. For the acellular vaccines, the frequency of adverse events was similar to that in the control (DT) group. CONCLUSIONS: The two acellular DTP vaccines we studied were safe, immunogenic, and efficacious against pertussis, whereas the efficacy of the whole-cell DTP vaccine was unexpectedly low.

Association of reactions after consecutive acellular or whole-cell pertussis vaccine immunizations.

OBJECTIVE: To evaluate the relative frequency of adverse reactions after initial and subsequent immunizations among infants receiving primary immunization with acellular (DTaP) or whole-cell (DTP) pertussis vaccine with diphtheria and tetanus combined. METHODS: We examined the occurrence of common reactions in 2127 infants within 48 hours after immunization at 2, 4, and 6 months with one of 13 DTaP or with Lederle DTP (WCL). Data on at least two consecutive immunizations were available for 357 WCL recipients and 1770 DTaP recipients. For these analyses, reactions evaluated included fever of 100.4 degrees F (38 degrees C) or greater, redness of 21 mm or larger, swelling of 21 mm or larger, moderate or severe pain, moderate or severe fussiness, loss of appetite, drowsiness, and vomiting. RESULTS: With one exception, reactions were approximately 1.5 to 8 times more likely to occur in WCL recipients if the same reaction had been observed at the previous immunization (the single exception was redness after the second immunization). Both initial and repeated reactions were less likely in DTaP than in WCL recipients. As with WCL recipients, risks of repeated reactions in DTaP recipients were higher than the risks of initial reactions (from 2.5 to 24 times as high). CONCLUSION: Reactions after a second or third immunization with either WCL or DTaP vaccine are more likely to occur in infants who had the same reaction after the preceding immunization. Absolute risks of repeated reactions tended to be lower after DTaP vaccine than after the WCL vaccine.

Comparison of five calculation modes for antibody ELISA procedures using pertussis serology as a model.

During a phase III pertussis vaccine trial, serum antibody responses were measured by two enzyme-linked immunosorbent assays (ELISA) for pertussis toxin and filamentous haemagglutinin. These were used both for studies of antibody levels after vaccination and for diagnostic purposes. Since the absorbance values obtained were not directly proportional to the amount of antibody in the samples, ELISA optical densities were transformed to units by calibration to a reference serum. Five different calculation modes were compared. In four of these modes unit calculations were based on the relationship between dose response curves of the serum sample and a reference serum. In addition, traditional endpoint titres were included in the comparison. The calculation mode using reference line units showed the highest reproducibility, with intrassay coefficients of variation (CV) within the same test plate of 4-7% and interassay CVs of 12-14%. The CVs among the other methods ranged from 6 to 31% for intra-assay comparisons and from 12 to 47% for interassay comparisons. Furthermore, the CV values for intra-assay variations were used to calculate standardized differences between 79 pairs of acute and convalescent sera from cases confirmed by culture. These differences were then used to estimate the 'diagnostic sensitivity' for the different calculation modes. The results indicated that use of the reference line units was the most sensitive, whereas use of the end point titers was the least sensitive of these calculation modes.

Similarity/equivalence trials for combination vaccines.

In a similarity or "equivalence" trial of a combination vaccine, we wish to show that the combination is sufficiently similar to the separately administered components in some measure of safety, immunogenicity, or efficacy to justify use of the combination. In this setting it is usually required to show similarity in one direction only; specifically, we design the trial to rule out superiority of the separate components by as much as a prespecified quantity theta 0 in an appropriate outcome measure (e.g., a difference or ratio). It is crucial that theta 0 be chosen to be clinically meaningful, so that any difference or ratio less than theta 0 is truly acceptable to clinicians. Estimation is generally more relevant than hypothesis testing in such a trial, and consequently it is natural to consider design and analysis in terms of confidence intervals. The same sample sizes can be obtained, however, from a hypothesis testing approach. The appropriate null hypothesis is that the separate components are superior to the combination by at least theta 0, with rejection of the hypothesis supporting a conclusion of similarity. It is not appropriate to design the trial to test the null hypothesis of no difference, as we would do if we wished to demonstrate superiority of the combination vaccine; failure to reject that hypothesis does not prove similarity, and we might reject the hypothesis when the true difference is unimportant clinically. Further, this inappropriate approach may result in a sample size either larger or smaller than necessary. Sample size formulations are available for various types of comparative measures, e.g., a difference of normally distributed means, a difference or ratio of proportions, and a ratio of hazards.

Outcome of the Vaginal Infections and Prematurity Study: results of a clinical trial of erythromycin among pregnant women colonized with group B streptococci.

OBJECTIVE: Our purpose was to determine whether erythromycin treatment of pregnant women colonized with group B streptococci would reduce the occurrence of low birth weight (< 2500 gm) and preterm (< 37 completed weeks) birth. STUDY DESIGN: In a double-blind clinical trial, 938 carriers of group B streptococci were randomized to receive erythromycin base (333 mg three times a day) or matching placebo beginning during the third trimester and before 30 weeks and continuing for 10 weeks or until 35 weeks 6 days of pregnancy. RESULTS: Pregnancy outcomes were available for 97% of randomized women; 14% of subjects withdrew from the trial. Birth weight < 2500 gm occurred in 8.6% of the erythromycin and 6.1% of the placebo recipients (relative risk 1.4, 0.9 to 2.2, p = 0.16). Preterm delivery occurred in 11.4% of women randomized to erythromycin and in 12.3% randomized to placebo (relative risk 0.9, 95% confidence limits 0.6 to 1.3, p = 0.65). Greater benefit of erythromycin in reducing these outcomes was not observed among women reporting the best compliance. CONCLUSIONS: In this study of pregnant women colonized with group B streptococci treatment with erythromycin was not shown to be effective at prolonging gestation or reducing low birth weight. Greater than anticipated complicating factors, including spontaneous clearance of the organism, use of nontrial antibiotics, and density of colonization, may have resulted in population sizes too small to detect a benefit of treatment. Future studies should take these factors into account in determining sample sizes.

Sample size and power for prospective analysis of relative risk.

In a placebo-controlled vaccine efficacy trial or a trial of equivalence of vaccines, one may wish to show that relative risk of disease is less than a specified value R0, not equal to one. This paper compares three methods for estimating relative risk in the binomial setting, based on a logarithmic transformation, likelihood scores, and a Poisson approximation. Exact power and size of test are calculated by enumeration of possible binomial outcomes, and power is approximated from asymptotic formulations. Although the score method is generally preferable, for most studies of practical interest the log and score methods are comparable, and the Poisson method is also appropriate for small risks, up to about 0.05. When true and null relative risks are less than one, unequal allocation of study individuals can increase power, and the asymptotic formula for the log method may substantially underestimate power; in such a study the power approximation for the score method is more reliable, even if the log method is used in analysis. Exact power calculations are helpful in planning studies. The log and Poisson methods, but not the score method, apply readily in the case of unequal follow-up.

The use of external monitoring committees in clinical trials of the National Institute of Allergy and Infectious Diseases.

Randomized clinical trials are being conducted and/or sponsored by all scientific divisions of the National Institute of Allergy and Infectious Diseases (NIAID). External committees to review the progress of ongoing trials and to make recommendations to the Institute concerning continuation or termination are an integral part of many of these trials. These committees have evolved considerably from the ad hoc committees, called together when a need arose, which were used beginning in the mid 1970s. Currently, there are many monitoring committees operating for NIAID-sponsored trials. They function in a variety of ways, based partially on historical precedent and partially on the specialized requirements of the particular trial; there is no 'standard operating procedure' for the Institute, or even for divisions within the Institute. One of the major issues faced in establishing the data and safety monitoring board for AIDS treatment trials was access to the meetings of this board and to the reports of interim data that the board reviewed. After much discussion, procedures were established that restrict such access to a very limited group of programme and statistical centre staff. These procedures, while remaining controversial, appear necessary to ensure confidentiality and the integrity of the clinical trials process.

Pertussis antibodies, protection, and vaccine efficacy after household exposure.

During a randomized trial of a cellular pertussis vaccines, significantly fewer recipients of a two-component vaccine (Japanese National Institute of Health [JNIH]-6) were diagnosed as primary or coprimary cases in households than either placebo recipients or those who received a monocomponent pertussis toxoid vaccine (JNIH-7). After household exposure to a culture-confirmed primary case, efficacy for JNIH-6 was estimated to be 35% (95% confidence interval, -14% to 57%) against any culture-confirmed disease and 58% (95% confidence interval, -6% to 84%) against clinical disease with 21 days or more of coughing spasms. The corresponding efficacy estimates for JNIH-7 were 67% (95% confidence interval, 32% to 80%) and 82% (95% confidence interval, 41% to 96%). Differences between the JNIH-6 and JNIH-7 vaccines in efficacy after household exposure were not statistically significant. No association could be established between protection against pertussis after household exposure and serum levels of IgG antibody to pertussis toxin or filamentous hemagglutinin in vaccinated individuals, in either study children or other household members.

Acellular pertussis vaccines. Efficacy and evaluation of clinical case definitions.

The efficacy of two acellular pertussis vaccines was estimated for various clinical case definitions, with and without the requirement of culture confirmation, from a randomized trial in Sweden. Efficacy increased with duration of coughing spasms and when the case definition included whoops or whoops plus at least nine coughing spasms a day. After deletion of clinical cases not believed to be caused by pertussis, efficacies were closer to the higher values for culture-confirmed disease. Nonspecificity of the clinical criterion "21 days of coughing spasms with whoops" resulted in estimates of predictive value for pertussis of 85% for placebo recipients and 56% for vaccinees. We conclude that laboratory confirmation of suspected cases is needed in pertussis vaccine trials. A suggested case definition is 21 days or more of coughing spasms with confirmation by culture, serologic study, or household exposure to culture-confirmed pertussis.

Antepartum cultures for Ureaplasma urealyticum are not useful in predicting pregnancy outcome. The Vaginal Infections and Prematurity Study Group.

To test the hypothesis that genital colonization with Ureaplasma urealyticum would predict adverse pregnancy outcome, 4934 women from five medical centers were evaluated for vaginal colonization with U. urealyticum between 23 and 26 weeks' gestation and followed up to delivery. U. urealyticum colonization was associated with maternal age, parity, racial-ethnic group, martial status, income, education, smoking, number of sexual partners, and colonization with Trichomonas vaginalis, Mycoplasma hominis, and bacterial vaginosis. After adjustment for medical and sociodemographic factors in a multivariate analysis, there was no difference in the mean birth weight or proportion of low-birth-weight infants delivered by women who carried U. urealyticum and those who did not. U. urealyticum colonization at 23 to 26 weeks was not associated with preterm rupture of membranes, preterm labor, or preterm delivery. A positive vaginal culture for U. urealyticum in midgestation does not predict those women at risk for preterm labor, preterm delivery, preterm premature rupture of membranes, or delivery of a low-birth-weight infant.

Value of a run-in period in a drug trial during pregnancy.

The Vaginal Infections and Prematurity (VIP) Study included a clinical trial of the efficacy of erythromycin in preventing adverse pregnancy outcomes. A prerandomization run-in period was part of the trial design. During this period, women were given 1 week's supply of placebo pills to test their compliance. Those who met certain criteria for compliance, were otherwise still eligible, and agreed to participate were then randomized to receive erythromycin or placebo for a maximum of 10 weeks. During 2 years of the VIP Study, 1476 (71%) of 2071 women who began the run-in period were randomized. Women at least 30 years of age, those not smoking during pregnancy or shortly before pregnancy, and those still working outside the home at the time of enrollment were more likely than other women to be randomized after the run-in. Calculations of relative efficiency, based on a standard sample-size formula, suggest that the VIP run-in procedure not only eliminated potentially noncompliant women before randomization but also increased the power of the trial. Similar calculations which incorporated costs suggest that the run-in also resulted in lower costs, compared to a trial with equivalent power but no run-in period.

Infections complicating low-risk cesarean sections in community hospitals: efficacy of antimicrobial prophylaxis.

A prospective study of women with low-risk cesarean sections was conducted in four community hospitals to determine the frequency of postoperative infections and identify factors predisposing to endometritis and wound infection. Low-risk cesarean section was defined as a scheduled procedure without an urgent indication, with any duration of ruptured membranes being less than or equal to 12 hours. In a cohort of 1863 patients, 26 (1.4%) developed endometritis and 21 (1.1%) had wound infections. Primary cesarean section was associated with endometritis in the cohort (p less than 0.01) and in a retrospective study with the same cases as in the cohort (p = 0.01). Absence of antibiotic prophylaxis was associated with endometritis (p less than or equal to 0.013) or endometritis with wound infection (p less than 0.01) in both studies. Without prophylaxis 37 such infections occurred in 957 (3.7%) women; with prophylaxis eight infections occurred in 906 (0.9%) women. Routine timely antibiotic prophylaxis in low-risk cesarean sections could lead to an annual national savings of approximately $9 million.

Unlikely association between clinically apparent herpesvirus infection and coronary incidence at older ages. The Framingham Heart Study.

Experimental studies in chickens have shown a relationship of a herpesvirus to atherosclerosis. The hypothesis of an association in humans was tested by using data on the history of cold sores and other manifestations of herpes infections reported by 658 male and 919 female participants (ages 58 to 89) in the Framingham Heart Study from 1977 to 1979 and on the prevalence and subsequent 6-year incidence of coronary heart disease (CHD). Approximately 40% of the men and 52% of the women reported a history of ever having "fever blisters or cold sores." Overall, there was no association between a history of such oropharyngeal manifestations and prevalent CHD. Only in the subgroup of women with recurrent infections was there a suggestion of a possible relationship (relative risk = 1.5, 95% confidence interval 1.0 to 2.1). Among members of the cohort without CHD at baseline there was no association between the history of cold sores, chicken pox, shingles, or infectious mononucleosis and 6-year CHD incidence. However, a possible interaction among women with recurrent herpes, lower levels of serum cholesterol, and incidence of angina pectoris without myocardial infarction was suggested in post hoc analyses. These data from the Framingham cohort do not support the notion that any self-reported clinically manifest herpesvirus infection has a strong etiological role in older persons, but they do raise issues to be addressed in any further research.